Oil production prediction plays a significant role in designing programs for hydrocarbon reservoir development, adjusting operations and making decisions. The accuracy of oil based on single methods is limited since more unconventional reservoirs are being exploited. Artificial intelligence technology data decomposition widely implemented multi-step forecasting strategies. In this study, hybrid model was proposed two-stage decomposition, sample entropy reconstruction long short-term memory...

Cognitive load recognition has been widely studied recently, but how to find the effective and robust feature representations from electroencephalography (EEG) signals is still a challenge. In this paper we design lightweight 1D 2D Convolutional Neural Networks (CNNs) with large-margin softmax loss functions for cognitive recognition. First, extract frequency domain features EEG signals. Then, extracted vectors matrices are used train CNNs. Our approach approximately achieves 93% accuracy...

Abstract It is common sense the wettability alteration in carbonate reservoirs stems from oil–rock interactions; however, exact oil component acting as dominant role remains unclear. This study adopts molecular dynamics technology, takes {101¯4} calcite crystal surface mineral of reservoir, uses multiple single-component oils to characterize real crude oil, and conducts research on reservoirs. First, based chromatographic analysis results a certain reservoir Middle East, simulated divided...

Summary Oilfield development performance prediction is a significant and complex problem in oilfield development. Reasonable of can guide the adjustment plan. Moreover, reservoir will change slowly during because flowing water however, previous networks that forecast production dynamics ignored it, which leads to inaccurate predictions. Routine well-wise injection measurements contain important subsurface structure properties. So, for dynamic oil/water two-phase waterflooded reservoirs, we...

Crude oil–aqueous solution interactions play an important role in low-salinity waterflooding, but their effects and mechanisms have not been well clarified. Core flooding experiments were first conducted to analyze the of crude on enhanced oil recovery (EOR). Then, interfacial tension (IFT) dilatational rheology (IDR), total organic carbon (TOC), ζ-potential measurements combined investigate waterflooding. results showed that are waterflooding adjusting ionic composition aqueous could affect...

There is not a consistent view about the mechanism of wettability alteration during low salinity water flooding. This paper highlights extensive studies to investigate on mineralogically different carbonates. Contact angle measurements were conducted characterize changes quantitatively. The results clearly revealed that carbonate rock surfaces can be altered more water-wet condition by lowering salinity. trend maximum change contact (MCCA) variation with dolomite/calcite content in fairly...

Low-salinity water flooding, known as an environmentally friendly and efficient oil recovery technology, has attracted the attention of several researchers all over world. However, its field application is suffering restrictions because ambiguous mechanisms by controlling salinity. In this study, a flooding microfluidic experiment was conducted to investigate pore-scale mechanism enhanced sweep efficiency low-salinity flooding. This used reservoir-on-a-chip that preserved real rock...

Due to the spatial network structure, heavy oil has a high threshold pressure gradient when it flows through porous media, and plays crucial role in distribution of remaining oil. In previous study, common methods measure include microflow-established differential (MFEDP) method, capillary equilibrium percolation curve fitting method. this sample from SZ36-1 oilfield was analyzed for basic physical properties based on comparison measurement study influence mobility then an independently...

Image segmentation of pear leaf diseases is a crucial method to scientifically determine the occurrence degree diseases. Pear sizes differ lot affected by period and degree. Convolutional neural network has fixed receptive field when extracting images, which makes it difficult fully extract features different sizes. At same time, continuous down-sampling reduces resolution feature map loses spatial information, resulting in rough mask image restored upsampling. To solve this problem, model...

This article presents an extensive relative permeability study for low salinity water flooding (LSWF) in minerally different carbonate rocks. The experiments were performed with unsteady state method using artificial cores, crude oil and synthetic brines. effect of rock composition on was obtained. results show that the (water) increases (decreases) first then decreases (increases) decrease Actually, there exists optimal (2384.6 mg/L) which maximizes minimizes permeability. When is kept...

Abstract The mechanism of the anisotropic permeability effects on well patterns and reservoir development are investigated by coordinate transformation, fluid flow analysis, concepts. Anisotropy has destruction reconstruction general patterns. originally designed flooding units broken up, new pattern made up wells that belong to different original units. behavior possesses strong randomness, leads chaotic relationship among injection production wells, uncertain productivity oil reservoirs....

ABSTRACT: As the greenhouse effect gradually intensifies, addressing CO2 as primary gas has become a pressing concern. Currently, predominant carbon sequestration technique involves storing in reservoirs or saline aquifers. However, readily dissolves water, forming carbonic acid, which can compromise geological formations and lead to stress corrosion cracking. Stress cracking (SCC), chemical activation mechanism, fluid adsorption on crack surfaces fluid-assisted propagation. While there is...